DataFrame doesn't read empty values from csv in case of only 1 column

[asmirnov82]

Test to reproduce:

[Fact]
public void TestSaveAndReadCsvWithOnlyOneColumnAndNullValues()
{
    using MemoryStream csvStream = new MemoryStream();
    var column = new SingleDataFrameColumn("Data", 3);
    DataFrame dataFrame = new DataFrame();
    dataFrame.Columns.Add(column);
    column[0] = 0;
    column[1] = null;
    column[2] = 2;

    DataFrame.SaveCsv(dataFrame, csvStream, header: true);

    csvStream.Seek(0, SeekOrigin.Begin);

    DataFrame readIn = DataFrame.LoadCsv(csvStream);

    Assert.Equal(column[0], readIn[0, 0]);
    Assert.Null(readIn[1, 0]);
    Assert.Equal(column[2], readIn[2, 0]);
}

In case of adding additional column with the same values - everything works fine

[feiyun0112]

The current design is to ignore the empty new line.

https://github.com/dotnet/machinelearning/blob/main/test/Microsoft.Data.Analysis.Tests/TextFieldParserTests.cs#L413

[asmirnov82]

In this case, I close the ticket. Thanks!